The National Colonoscopy Study is the only trial that provides a direct randomized comparison of the performance of a single screening colonoscopy with programmatic (annual) fecal occult blood testing (FOBT) utilizing a sensitive slide in the asymptomatic general population. In addition, the outcome of this study can provide measurements of the relative impact on incidence and mortality of these two screening strategies using mathematical modeling. These data are unavailable from any ongoing or planned long term randomized controlled trials (RCTs). When colonoscopy was introduced as a screening option in the 1997 GI Consortium CRC guidelines, it was stated that this screening modality was potentially so powerful that strong data needed to be obtained to support its widespread clinical application, especially compared to the commonly used screening test, the FOBT. The data obtained in this study will be of value in clinically prioritizing between direct screening colonoscopy and FOBT directed colonoscopy. The justification for this application is that we need a minimum of 3 rounds of FOBT screening in all subjects in the FOBT arm in order to adequately compare programmatic (annual) FOBT with a single colonoscopy. This will require support for three additional years and will permit two more rounds of FOBT screening with assessment for interval cancers. We have successfully accrued our goal of 3500 subjects;completed screening colonoscopy in the colonoscopy arm subjects;completed baseline FOBT screening for all in the FOBT arm;and completed a second and third round of FOBT screening for a proportion of the subjects in the FOBT arm. The infrastructure for this RCT is well organized and established including the clinical centers, pathology review center, the coordinating center, the FOBT centralized laboratory, and biorepository. The specific aims of the proposed study are to continue the FOBT screening rounds;determine the adherence per round of FOBT and the characteristics of those adhering;and evaluate the cumulative percent of subjects with significant clinical findings with a program of annual FOBT compared to that of a screening colonoscopy. We will use a microsimulation model (MISCAN) to assess long term effectiveness, costs, and cost-effectiveness of the two screening strategies using the specific levels of adherence to screening and clinical findings of this study. In addition, we will continue adding specimens from the FOBT directed colonoscopy subjects to the biorepository established primarily from subjects in the colonoscopy arm. This study will have a direct impact on current CRC screening guidelines by filling the gaps in our knowledge of the magnitude of reduction in incidence and mortality of screening colonoscopy compared to programmatic FOBT through mathematical modeling of RCT data.